DESCRIPTION: The investigator proposes to continue his investigations on the mechanism and purpose of DNA methylation using Neurospora as a model system. In prior studies, the investigator showed that duplicated sequences in Neurospora, including a duplicated 5S RNA gene and a 5S pseudogene, are heavily altered by point mutations during the sexual phase of the life cycle. This process, called RIP (for repeat-induced point mutation), leads to methylation of the RIPed sequences. These observations led to studies conducted during the last funding period regarding the extent, nature, and purpose of DNA methylation in Neurospora. The overall objective of the current proposal is to understand the function and control of DNA methylation in a model eukaryote. The specific aims include the following: (1) to determine why DNA methylation-deficient mutants become aneuploid, (2) to isolate new DNA methylation mutants by approaches that will facilitate cloning the genes, (3) to isolate and characterize the genes involved in DNA methylation, (4) to define the signals for de novo methylation, (5) to identify and purify proteins that bind specifically to methylated sequences, and (6) to explore the potential connection between DNA methylation and DNA replication. The proposal springs from evidence linking methylation in genetic and epigenetic phenomena such as imprinting. Many correlations between methylation and the lack of gene activity have accumulated and there is strong evidence that methylation prevents gene expression. Findings that eukaryotic methyl transferases preferentially methylate hemi- methylated CpGs have provided support for the models proposed by Riggs and Holliday that methylation is propagated through an enzyme that methylates symmetrical sites. However, it has become clear in studies on fungi that propagation of methylation can occur at non-symmetrical sites. In Neurospora, methylation can take place de novo and in a manner that does not propagate the methylation pattern. About 1.5% of the C's of N. crassa DNA are methylated, but the bulk of the genome is devoid of methylation. Three methylated sequences characterized in detail include the tandemly arranged rDNA, the 1.6 kb zeta-eta region, and the psi region, the latter being relics of RIP, repeat induced point mutation. This is a process discovered by Dr. Selker to be mechanism that detects duplicated sequences in the haploid nuclei of dikaryotic cells after fertilization. Both copies of the duplicated sequence then become frequent targets of G:C to A:T mutation, so much so that 30% of G:C pairs can be altered. Frequently, sequences altered by RIP become methylated. Methylation of RIPed sequences occurs in vegetative cells inactive for RIP and sequences altered by RIP become methylated de novo when reintroduced by transformation. Thus, RIP can change unmethylated chromosomal sequences into targets for methylation. The methylation can extend beyond the mutated region and beyond the boundaries of the segment that was originally duplicated. Methylation is not limited to symmetrical sites. Sequences altered by RIP become methylated but not their nonmethylated counterparts, when reintroduced by transformation. Thus, signals for methylation created by RIP are portable. RIPed sequences become remethylated at all chromosomal positions tested. A question of general interest is how are signals for methylation generated by mutations from RIP.